summary of roundtable on online advertising - oecd · however, online advertising, especially...

Organisation for Economic Co-operation and Development

DSTI/CP(2018)12/FINAL

Unclassified English - Or. English

13 August 2018

DIRECTORATE FOR SCIENCE, TECHNOLOGY AND INNOVATION

COMMITTEE ON CONSUMER POLICY

Summary of Roundtable on Online Advertising

JT03434988

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the

delimitation of international frontiers and boundaries and to the name of any territory, city or area.

2 │ DSTI/CP(2018)12/FINAL

SUMMARY OF ROUNDTABLE ON ONLINE ADVERTISING Unclassified

This document summarises a Roundtable on Online Advertising that was held by the

Committee on Consumer Policy (CCP) on 18 April 2018. It highlights some of the key themes

and issues raised during the discussions. The final agenda for the roundtable is attached as an

annex.

This paper was approved and declassified by written procedure on 31 July 2018 and prepared

for publication by the OECD Secretariat.

© OECD 2018

You can copy, download or print OECD content for your own use, and you can include

excerpts from OECD publications, databases and multimedia products in your own

documents, presentations, blogs, websites and teaching materials, provided that suitable

acknowledgment of OECD as source and copyright owner is given. All requests for

commercial use and translation rights should be submitted to [email protected].

DSTI/CP(2018)12/FINAL │ 3


Summary of Roundtable on Online Advertising

Advertising is not new. Indeed, it is as old as commerce itself. However, online advertising,

especially “online behavioural advertising”, which uses personal data to target ads, is

relatively new. It is now big business, having recently overtaken television to be the dominant

form of advertising worldwide in terms of expenditure.

The supply chain and business models associated with online advertising are much more

complex than those associated with more traditional forms of advertising. Further, the ability

to personalise online advertising using personal data raises new benefits and risks for

consumers. Given this, it is important that consumer authorities understand how online

advertising works, what the potential benefits for consumers are, and also, the likely risks. In

order to explore these questions, the Committee on Consumer Policy held a roundtable on 18

April 2018 with panellists from the private sector, academia and consumer protection

authorities.

Following a brief introduction and welcome, the roundtable comprised two sessions. The first

session focussed on current trends in online advertising, providing an overview of some of

the technologies used to facilitate online advertising, and an industry perspective on some of

the potential benefits and risks of online advertising. The second session then focussed more

on the identification of risks for consumers, and potential policy responses from consumer

protection authorities. This roundtable summary highlights some of the key themes and issues

raised during the roundtable. The final agenda is attached as an annex.

Main points

Consumers can benefit from services funded by targeted advertising

Online advertising provides a range of potential benefits and risks for consumers. There are

two key ways in which online advertising can benefit consumers. The first is that it can fund

the provision of “free” online services to consumers. For example, online advertising revenue

allows for internet search and social media services to be provided to consumers without

monetary charge. The second is that, through the use of personal data, online advertising can

allow for better targeted online advertising. This has the potential to reduce consumer search

costs, to introduce consumers to more relevant products or services, and to highlight

promotions to consumers; all of which could be beneficial for consumers.

But online ads can mislead, especially if consumers cannot identify them

However, online advertising also presents potential consumer risks and challenges. Some of

these are not new. For example, the ability for advertising to mislead consumers, either

through unfounded claims, omissions of key terms or limitations, or misrepresentations

regarding the nature of the material (i.e. as unbiased opinion rather than commercial content),

is not unique to online advertising but remains a problem both on and offline. Identification

of advertising material can be particularly challenging for consumers in relation to newer

forms of online advertising such as native advertising and paid endorsements on social media

(i.e. “influencer marketing”). In this respect, there appears to be potential to improve

disclosures. In particular, behavioural experiments show promise in terms of improving online

advertising disclosures in various contexts.



Consumers may also be more prone to manipulation and face privacy and security

threats

Newer issues emerge from online behavioural advertising, which uses personal data to

personalise and target advertising. In particular, if this is used to target behavioural biases or

consumer vulnerabilities, it could potentially be used to manipulate consumer decision

making. It could also potentially be used to implement personalised pricing and other forms

of discrimination between consumers. However, it may be difficult to identify whether and

how online advertising is being used in this way . For this reason, this is an area that could

potentially benefit from further research.

The widespread collection of personal data that is used to inform online behavioural

advertising also has implications for consumer privacy and security. In most OECD countries,

consumer agencies do not bear primary responsibility for these issues, but they may have a

role in ensuring that such practices (and the disclosure of such practices) are not deceptive or

unfair

There is increasing industry awareness about the potential adverse impacts of

some online advertising practices on consumer trust

Industry is starting to recognise the concerns that drive consumers to take steps to block

unwanted online advertising. In particular, many consumers are frustrated when online

advertising negatively impacts their online experience, either by slowing down services or

interrupting them. Consumer trust issues raised by privacy and security concerns are also of

key importance. For this reason, a number of industry players are designing their services in

ways that maintain consumer privacy and security without compromising on the ability to use

consumer data at an aggregate level to constantly improve services and raise revenue.

Session 1: Current trends and industry perspective

The roundtable opened with a brief introduction and welcome from the Chair of the

Committee on Consumer Policy (CCP), Mr Hugh Stevenson.

Session 1 began with Mr. Jules Polonetsky, CEO of the Future of Privacy Forum, providing

an overview of the online advertising industry, focussing on how and why consumer

behaviour is tracked online. He noted that understanding and measuring ad effectiveness is

the main driver for tracking consumers online. He then briefly outlined the supply chain and

technologies underpinning this tracking. Mr. Polonetsky noted that cookies (small pieces of

information saved as a text file on a user’s device) were traditionally used to track consumers

online but have become less effective since consumers now use a variety of devices and

mobile apps do not support cookies. He noted some other measures being used to track online

behaviour, including authenticated sign on, the IP address or local router SSID accessed by

the device, browser “fingerprinting”, and geo-location data. He also noted examples of

password managers being used to track consumers online (see, e.g. Acar (2017[1]) and

Brandom (2017[2])). Mr .Polonetsky then highlighted some of the ways that the private sector

is responding to concerns about tracking including by reducing the collection and use of

personally identifiable information, changing default settings for data collection and use, and

allowing consumers to choose settings that limit ad tracking. He also noted that the European

Union’s General Data Protection Regulation (GDPR) may offer further protections for

consumers.



Next, Ms. Aibhinn Kelleher, responsible for Product Policy Management at Facebook,

outlined Facebook’s approach to online advertising. She noted that over 5 million advertisers

are active on Facebook globally and that it is the advertisers, rather than Facebook, that are

responsible for ensuring compliance with applicable laws. Ms. Kelleher showed how

Facebook ads are disclosed, namely, with the word “Sponsored” displayed in the top left hand

side of the ad. While compliance with relevant laws is the responsibility of advertisers,

Facebook has a number of policies that govern how advertising works on Facebook. These

include restrictions on certain types of ads, including those for alcohol; gambling and lotteries;

pharmaceuticals; dating; software; subscription services; weapons and explosives; and

tobacco. In addition, the policies require: transparency about the product or service being

advertised; that ads are not false, misleading or deceptive; and that advertisers comply with

relevant laws. In terms of enforcement, Facebook allows for users to report a violation of the

Facebook Terms of Service. This can result in Facebook reviewing, removing or blocking an

ad that does not comply with their policies. They also work with government authorities that

contact them regarding potentially unlawful ads. In addition, prior to publishing, ads are

reviewed either by machine based software or by staff for compliance with Facebook’s

policies. Ms. Kelleher noted that the same standards apply to ads on Instagram.

Session 1 finished with Ms. Catherine Armitage, Head of Digital Policy at the World

Federation of Advertisers (WFA), providing an industry perspective on online advertising.

Ms. Armitage noted that worldwide expenditure on online advertising has now surpassed

spending on other forms of advertising. She also noted the increasing use of ad blocking

technologies, representing around 11% of internet users in late 2016, and growing at a rate of

30% each year. She highlighted some of the problems faced by members of the WFA in

relation to online advertising including supply chain complexity; ad fraud; threats to brand

reputation and consumer declining sentiment. Ms. Armitage provided a diagrammatic

illustration of the supply chain to highlight its complexity, and noted the number of players

involved (often over 50 for each ad) and the fact that each takes a cut of the money. She noted

that advertisers can be subject to fraud where they are charged for ads that are never seen by

consumers noting that some of these schemes can be quite elaborate. She also noted that

brands can suffer reputational damage if adequate controls are not in place to manage where

ads are published. Further, she noted concerns around consumer sentiment in relation to online

advertising. She highlighted the need to ensure that ads do not overly interfere with the

consumer’s online experience and the importance of consumer trust in relation to the

collection and use of personal data.

The discussion that followed covered a range of issues. There was discussion about the use of

Artificial Intelligence (AI) in advertising and questions about whether guidance was needed.

Industry representatives noted that while machine learning was and had been used in online

advertising for some time, more sophisticated forms of AI were not being used extensively

and so guidelines may be premature. There were questions as to how platforms deal with

differences in legislation between jurisdictions and whether enforcement is undertaken on an

ex ante or ex post basis. Ms. Kelleher noted that the responsibility is largely with the

advertisers but that Facebook works with relevant authorities to remove offending ads as

required. She also noted that Facebook undertakes some pre-screening of ads but like

anything, it is not perfect. She noted that work with enforcement agencies is largely reactive

rather than proactive. In relation to trends in supply chain complexity, Ms. Armitage noted

that some advertisers are trying to simplify their supply chains. It was also noted that GDPR

may lead to some actors withdrawing from Europe. There was also discussion about the role

of self-regulation with some noting that self-regulation seems to work better for misleading

ads but less well for issues relating to overall consumer “annoyance” with and mistrust of



online advertising. The role of the private sector in bringing technological solutions to market

was also noted.

Session 2: Identifying and addressing the risks

Session 2 started with a presentation from Ms. Laura Sullivan, Senior Attorney, Division of

Advertising Practices, United States Federal Trade Commission (FTC).1 Ms. Sullivan

discussed the issue of consumer identification of online advertisements, drawing on analysis

undertaken by the FTC in 2014 and 2015, and published in 2017 (see FTC (2017[3])).2 The

analysis considered whether consumer identification of online native ads and search ads could

be improved with changes to the design of disclosures. The experiment made a number of

changes to the disclosures, including the language used to identify the content as advertising;

the placement of the text disclosure; the font size and colour of the text disclosure; and a

number of other visual cues (such as shading the background of the ad). The experiment tested

48 participants on their recognition of eight different sets of webpages containing ads (of

those, there was an equal split between search ads and native ads; desktop and mobile pages;

and original ads and modified ads; with the order varied by participant). Recognition was

subjectively measured based on participants’ spontaneous comments while interacting with

the webpages and their responses to a series of open-ended questions. Except for the final set

of webpages presented, participants were not asked whether anything on the webpages was

advertising participants’ eye movements and fixations on different areas of the screen were

also tracked. However, despite observing that participants spent significantly less time

looking at ads with the modified disclosures, no correlation was found between the likelihood

that a participant recognised an ad as an ad and the amount of time spent looking at it.

For both search ads and native ads, across the ads tested, ad recognition was found to improve

with the modified disclosures. Although the study did not test the effect of any one particular

disclosure change on ad recognition, it appears that certain disclosure techniques were

particularly effective including:

using a distinctive background shading alongside sufficiently noticeable ad labels

ensuring that the text colour sufficiently contrasts with the background colour (for

example, white text on a dark background is difficult to read on desktop)

locating disclosures close to an ad’s headline (but outside of “blind-spots” or where

readers tend not to look on webpage including in the right-hand corner of a web

page or a grouping of more than one ad).

A pre-recorded video of Mr. Ryan Calo, Associate Professor of Law from the University of

Washington, was played next. Mr. Calo discussed some of the ways that online advertising

can target consumer behavioural biases, potentially to the detriment of consumers. In

particular, he noted that there appeared to be a move from online ads that match consumers

with their preferences to those that match consumers with their “persuade-ability”. For

example, if an advertiser knows that a consumer is particularly influenced by scarcity then it

could target the consumer with ads that highlight the scarcity of the product. While such

conduct may appear relatively benign, it could be problematic if it preyed on a consumer’s

biases or vulnerabilities. Mr Calo also raised concerns about personal data being used to infer

a consumer’s reservation price, or maximum willingness to pay. While this might not be

inefficient, he noted, it may not be fair, especially if coupled with behaviour that preys on

consumer vulnerabilities (e.g. numbers dyslexia). Mr. Calo noted his work on what he termed

“digital market manipulation” and the potential to manipulate consumers in real time (see

Calo (2014[4])).3 He also noted work he had done with Alex Rosenblat on manipulation in



peer platform markets in a report called “The taking economy”, which could be especially

problematic given the information the platform has on users on both sides of the transaction

(see Calo and Rosenblat (2017[5])). He gave an example of research Uber has undertaken on

seeing whether or not consumers would be more willing to pay surge (i.e. higher) prices if

their phone battery is low. While recognising that the scope of the CCP is limited to consumer

issues, he also noted concerns about tactics employed in respect of consumers being used in

other contexts, for example, to influence political discourse.

Mr. Sam Wilson, Compliance Executive, Committee of Advertising Practice (CAP) (UK),

then discussed the role of CAP and its sister organisation, Advertising Standards Authority

(ASA) in policing and sanctioning misleading online advertising in the United Kingdom. In

particular, ASA/CAP is a self-regulatory body that is responsible for ensuring ads comply

with relevant laws and standards, including relevant ASA/CAP standards. Since 2011,

ASA/CAP’s remit has included online ads and since 2013, online behavioural ads. Each year

ASA/CAP receives around 20 000 to 30 000 complaints, with 90 % of these coming from

consumers. As a result, approximately 6 000 ads are withdrawn or amended annually due to

breaches of the code. In addition, he noted that ASA/CAP provide approximately 300 000

pieces of advice each year in respect of online ads. Mr. Wilson noted that ASA/CAP is

currently undertaking research on the recognition and labelling of online advertisements. It

has just concluded a call for evidence with the aim to use this to inform guidance on how to

produce effective online ad disclosures.

The last presentation was from Mr. Jason Novak, who works in User Privacy at Apple. Mr.

Novak discussed the ways in which Apple has implemented technological solutions to better

protect consumers against the risks of online advertising; in particular, to maintain consumer

privacy online. Mr. Novak noted Apple’s focus on consumer privacy in all aspects of its

business, and its belief that privacy is a fundamental right. However, he noted that there can

be trade-offs between functionality and privacy, and that some consumers want to access

services for free, meaning online ads are inevitable. Mr. Novak noted the disclosure efforts

taken by Apple to inform consumers about their privacy when using Apple products and

services; and to allow consumers to manage their privacy settings. He noted the heavy reliance

on third-parties in the online ecosystem. For example, for the top 100 websites in America,

on average, 16 third parties are involved in loading a website. In general, consumers will have

no ability to know how many third parties are collecting information on them when they go

online. While it is relatively easy to block all cookies online, Mr. Novak noted that this would

undermine usability and business models that rely on online advertising revenue (i.e. “free”

online services). Blocking of third-party cookies, however, retains most of the functionality

without undermining privacy. Hence, since 2003, Apple has blocked third-party cookies as

the default in its internet browser, Safari. More recently, Apple has introduced “Intelligent

Tracking Prevention” alongside “Storage Access API” to further restrict cross-site tracking

while maintaining functionality in the form of authenticated services. Mr. Novak noted that

Apple would like to see such approaches adopted as standards across the online environment

but that consensus is hard to reach.

A number of delegates asked questions about, and commented on, the FTC study in the

discussion that followed. In particular, support was expressed for undertaking experiments to

understand what forms of ad disclosures are most effective. Some delegates noted they had

been undertaking similar experiments on this issue and on other topics such as disclosure of

terms and conditions.4 There was interest from some delegates in undertaking “real life”

experiments, partnering with industry, if possible. In response, WFA and Facebook noted that

they were also undertaking similar types of experiments but that this was not usually done in

partnership with government agencies. Ms. Sullivan noted an academic study on native ad



disclosures that was undertaken in partnership with a restaurant review site (Zomato) (see

Shani and Nair (2016[6])) but noted that a limitation of this study was that it analysed only

changes in subjects’ behaviour (clicking and purchasing behaviour) and did not test for

consumer understanding. There was also some discussion about whether ad recognition is the

key issue or whether such experiments should also look at the impact of ads on consumer

behaviour.

In relation to “digital market manipulation”, delegates were interested to know whether and

to what extent these practices are actually occurring, and how such practices can be detected

(including personalised pricing). Mr. Calo noted that indeed there is little transparency around

this and academics and regulators are always playing catch-up in terms of understanding what

businesses are actually doing. He suggested this is an area that governments alongside

academia should continue to research. He noted that the key thing is to consider the incentives

and business models of the relevant businesses. He also noted that not all businesses are

engaged in these forms of behaviour and that businesses that prioritise privacy, such as Apple,

Microsoft and Mozilla, are encouraging.



References

Acar, G. (2017), No boundaries for user identities: Web trackers exploit browser login

managers, Freedom to Tinker, https://freedom-to-tinker.com/2017/12/27/no-boundaries-for-

user-identities-web-trackers-exploit-browser-login-managers/ (accessed on 26 April 2018).

[1]

Brandom, R. (2017), Ad targeters are pulling data from your browser’s password manager, The

Verge, https://www.theverge.com/2017/12/30/16829804/browser-password-manager-

adthink-princeton-research (accessed on 26 April 2018).

[2]

Calo, R. (2014), “Digital Market Manipulation”, The George Washington Law Review,

Vol. 82/4, http://www.gwlr.org/wp-content/uploads/2014/10/Calo_82_41.pdf (accessed on

30 January 2018), pp. 995-1051.

[4]

Calo, R. and A. Rosenblat (2017), “The Taking Economy: Uber, Information, and Power”,

Columbia Law Review, Vol. 117, http://dx.doi.org/10.1111/soc4.12493.

[5]

FTC (2017), Blurred Lines: An Exploration of Consumers' Advertising Recognition in the

Contexts of Search Engines and Native Advertising, FTC Staff Report, Washington,

https://www.ftc.gov/system/files/documents/reports/blurred-lines-exploration-consumers-

advertising-recognition-contexts-search-engines-

native/p164504_ftc_staff_report_re_digital_advertising_and_appendices.pdf (accessed on

18 December 2017).

[3]

FTC (2013), Blurred Lines: Advertising or Content? – An FTC Workshop on Native Advertising,

https://www.ftc.gov/news-events/events-calendar/2013/12/blurred-lines-advertising-or-

content-ftc-workshop-native (accessed on 26 April 2018).

[7]

Hanson, J. and D. Kysar (2000), “Taking Behavioralism Seriously: A Response to Market

Manipulation”, Roger Williams University Law Review, Vol. 6/1,

https://docs.rwu.edu/rwu_LR/vol6/iss1/8 (accessed on 26 April 2018).

[10]

Hanson, J. and D. Kysar (1999), “Taking Behavioralism Seriously: Some Evidence of Market

Manipulation”, Harvard Law Review, Vol. 112/7, pp. 1420-1572,

http://dx.doi.org/10.2307/1342413.

[8]

Hanson, J. and D. Kysar (1999), “Taking Behavioralism Seriously: The Problem of Market

Manipulation”, New York University Law Review, Vol. 74,

http://www.nyulawreview.org/sites/default/files/pdf/NYULawReview-74-3-Hanson-

Kysar.pdf (accessed on 26 April 2018), pp. 630-749.

[9]

Sahni, N. and H. Nair (2016), Native Advertising, Sponsorship Disclosure and Consumer

Deception: Evidence from Mobile Search-Ad Experiments, FTC,

http://ssrn.com/abstract=2737035 (accessed on 26 April 2018).

[6]



Annex A: Agenda for the Roundtable on Online Advertising

Introduction

Advertising has been around as long as commerce itself. While advertising has always had

the potential to raise consumer issues, online advertising has the potential to raise a whole

new set of risks and challenges for consumers. In particular, the tracking of consumer activity

online along with the emergence of big data and artificial intelligence (AI) present new

opportunities for marketers to target consumer biases at almost any time of the day and in any

context, as consumers spend more and more time online. Further, personalised advertising,

with its range of potential benefits and risks, can now be served up at scale.

Online advertising brings consumer benefits by funding access to content and services without

monetary payment. It also enables advertisements to be more relevant to consumer interests,

saving consumers search time and money. It brings potential risks as well. In particular, the

ability of online advertising to potentially target consumers’ behavioural biases at scale, and

to potentially tailor them to a consumer’s specific vulnerabilities, means that consumer

decision making may be more prone to manipulation than with other forms of advertising. In

addition, online advertising may present new ways in which to mislead consumers regarding

the full costs of a product or service, or in respect of unexpected terms and conditions of a

sale. This may be of particular concern where the advertising is targeted at children or other

vulnerable or disadvantaged consumers, as highlighted in the OECD E-commerce

Recommendation. Consumers may also risk being misled if they cannot easily identify online

content as advertising material. This was also raised as a concern in the E-Commerce

Recommendation. This is likely to be a particular concern with native advertising and user-

generated content such as “influencer” marketing, for example. Other threats from online

advertising could compromise consumer privacy and security, including through malicious

advertising (so called "malvertising").

As part of the CCP's continued work on behavioural insights and the protection of digital

consumers, this roundtable will examine these emerging trends and issues around online

advertising. The first half of the roundtable will provide a primer on trends in online

advertising from a number of perspectives. It will examine developments in "ad tech" that use

AI and machine learning algorithms to target advertising in real time based on personal data.

It will also provide an overview of the key types of online advertising and the new issues these

raise for consumers. In addition, the session will provide a business perspective on what

practices they put in place to best manage these risks for consumers. The second session will

then focus on the identification of the key risks for consumers, and potential ways to address

these. In particular, it will take an initial look at some of the problems being experienced by

consumers and some of the measures being put in place to protect consumers online where

constant exposure to advertising is now the norm.

Session 1: Current trends and industry perspective (14:15 – 15:15)

This session will provide an overview of the current trends in online advertising, focussing

on the potential benefits and risks to consumers. In addition, it will provide an industry

perspective on how to ensure that online advertising brings benefits to consumers, and

minimises the potential risks.



Panellists

Mr. Jules Polonetsky, CEO of the Future of Privacy Forum, will provide an overview of the

online advertising industry, with a brief overview of the potential benefits and costs to

consumers.

Ms. Aibhinn Kelleher, Product Policy Management at Facebook, will provide a business

perspective, outlining Facebook’s approach to online advertising and what it sees as the key

opportunities and risks for consumers, including an overview of the processes they have put

in place to protect consumers.

Ms. Catherine Armitage, World Federation of Advertisers, will provide an industry

perspective on the recent developments in online advertising, including the benefits and risks

for consumers, and the roles and responsibilities of industry players and policymakers in

ensuring that consumers can get the benefits from online advertising while minimising the

risks.

Box .A1. Discussion Questions

What are the key trends shaping the online advertising industry?

What risks do online advertising pose for consumers?

Which players create these risks and what do they do to manage them?

Who is best placed to deal with these risks?

Session 2: Identifying and addressing the risks (15:15 – 16:30)

This session will look at some of the key policy issues for consumers relating to online

advertising. It will look at what is required to ensure that consumers are not misled in their

dealings with online advertising, either due to an inability to identify online advertising, or

misrepresentations regarding the collection, use and sharing of personal information

associated with online advertising.

Panellists

Mr. Ryan Calo, Associate Professor of Law, University of Washington, will discuss (via

video) some of the ways that online advertising can target consumer behavioural biases and

some of the other consumer threats from online advertising, including in respect of the

collection of personal data.

Ms. Laura Sullivan, Senior Attorney, Division of Advertising Practices, US Federal Trade

Commission, will discuss the problems consumers may experience in identifying online

advertisements, drawing on quantitative analysis undertaken in relation to advertising

disclosures.



Mr. Sam Wilson, Compliance Executive, Committee of Advertising Practice (CAP) (UK),

will discuss CAP’s experience in policing and sanctioning misleading advertising online,

including some preliminary insights from a recent call for evidence on the recognition and

labelling of online advertisements.

Mr. Jason Novak, User Privacy, Apple, will discuss ways in which Apple has implemented

technological solutions to better protect consumers against the risks of online advertising (e.g.

through their internet browser, Safari), the motivations for this, and more generally, how

technological solutions can complement other measures to protect consumers against the risks

of online advertising.

Box .A2. Discussion Questions

Do generic protections against false and misleading information provide sufficient protection

for consumers in respect of online advertising?

If not, are additional measures required? If so, what are these? Should these involve self-

regulation, co-regulation, softer policy tools or new regulated requirements? What is the role

for consumer education?

What could policy makers do to support the evidence base to assess these questions?

Is the private sector likely to develop solutions? If so, how can policy best support this?

1 The order of Session 2 was changed due to technical difficulties in playing Mr. Ryan Calo’s video.

2 This analysis built on the findings of a 2013 FTC workshop on online advertising disclosure (see

FTC (2013[7])).

3 Mr. Calo noted the work of Hanson and Kysar in the 1990s around “market manipulation”, which

has influenced his work on such behaviour in the online context. See, e.g. Hanson and Kysar

(1999[8]); Hanson and Kysar (1999[9]) and Hanson and Kysar (2000[10]).

4 For example, the European Commission is expecting to publish work on online ad disclosures in

the coming months.

summary of roundtable on online advertising - oecd · however, online advertising, especially...

Documents